Path: utzoo!attcan!uunet!jarthur!usc!julius.cs.uiuc.edu!psuvax1!rutgers!rochester!pt.cs.cmu.edu!dsl.pitt.edu!pitt!willett!ForthNet
From: ForthNet@willett.pgh.pa.us (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: ANS TC Magnet for Vocabularies and :
Message-ID: <1958.UUL1.3#5129@willett.pgh.pa.us>
Date: 5 Nov 90 03:06:42 GMT
Organization: String, Scotch tape, and Paperclips.  (in Pgh, PA)
Lines: 56

Category 10,  Topic 14
Message 25        Sun Nov 04, 1990
B.RODRIGUEZ2 [Brad]          at 15:02 EST
 
> I believe that [implementing two different vocabulary mechanisms]
 > is what the vendors have in mind Brad.

Thanks for the clarification, Dennis.  Not that I agree, mind you.

> The other consideration is the behavior of word lists defined by
 > vocabulary.

Agreed.  I noted this in an early message; but until Dennis' reply (above) I
didn't see how the TC expected WORDLIST to fix the problem.

While we're on the subject, perhaps someone can comment on a statement I made
recently to my local FIG group, namely, that all vocabulary "architectures"
currently in use are subsets or special cases of a (FIG-style) tree structure,
provided that branches of the tree can be SEALed.  E.g.

    polyForth:  the trivial tree which has 8 branches from an
    empty root.

    F83 with ONLY/ALSO:  commonly, all vocabularies are branches
    from the ROOT vocabulary, although I believe branches may have
    sub-branches.  (I've never tried.)

Note that the architecture of vocabularies (how they appear to be organized)
does not say anything about the search order.  I'll now take a bigger risk and
posit that all search order mechanisms can be emulated with ONLY/ALSO.  E.g.

    polyForth:  each VOCABULARY name specifies a search order
    with one to four wordlists, emulated by
        : voc-name   ONLY list1 ALSO list2 ALSO list3 ALSO list4 ;

    fig-Forth:  each VOCABULARY name specifies its parent's search
    order to follow its own wordlist, as
        : voc-name   ONLY parent-voc-name ALSO this-word-list ;
    assuming that all vocabularies are sealed.  If branches of the
    tree are not sealed, fig-Forth is the trivial case.

I defer to the X3J14 team in this example's nomenclature, and refer to words
which set the search order "vocabularies," and words which refer to a single
list of words as "wordlists."  We really do need two terms here.

Does anyone know of a counterexample to these observations?

I realize that I'm making the TC's case for WORDLIST here.  Although I'm not
sure a word "VOCABULARY" can be defined for all emulations; I've defined
emulated vocabularies as colon definitions in these examples, and I think this
may be unavoidable for some emulations.

- Brad
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: dwp@willett.pgh.pa.us or uunet!willett!dwp